97 lines
3.0 KiB
Python
97 lines
3.0 KiB
Python
|
from pathlib import Path
|
||
|
|
|
||
|
|
from fastapi import APIRouter, Request
|
||
|
|
from fastapi.templating import Jinja2Templates
|
||
|
|
|
||
|
|
from app.db import fetch_all, fetch_one
|
||
|
|
|
||
|
|
router = APIRouter(prefix="/documents", tags=["documents"])
|
||
|
|
BASE_DIR = Path(__file__).resolve().parent.parent.parent
|
||
|
|
templates = Jinja2Templates(directory=str(BASE_DIR / "templates"))
|
||
|
|
|
||
|
|
|
||
|
|
def _iso_rows(rows: list) -> list:
|
||
|
|
for row in rows:
|
||
|
|
for key, val in list(row.items()):
|
||
|
|
if hasattr(val, "isoformat"):
|
||
|
|
row[key] = val.isoformat()
|
||
|
|
return rows
|
||
|
|
|
||
|
|
|
||
|
|
@router.get("")
|
||
|
|
async def documents_page(request: Request):
|
||
|
|
summary = {
|
||
|
|
"documents": 0,
|
||
|
|
"total_words": 0,
|
||
|
|
"unique_words": 0,
|
||
|
|
"avg_sentiment": 0.0,
|
||
|
|
"positive": 0,
|
||
|
|
"neutral": 0,
|
||
|
|
"negative": 0,
|
||
|
|
}
|
||
|
|
try:
|
||
|
|
row = fetch_one(
|
||
|
|
"""
|
||
|
|
SELECT COUNT(*) AS docs,
|
||
|
|
COALESCE(SUM(word_count), 0) AS words,
|
||
|
|
COALESCE(AVG(sentiment_compound), 0) AS avg_sent
|
||
|
|
FROM document_analytics
|
||
|
|
"""
|
||
|
|
)
|
||
|
|
if row:
|
||
|
|
summary["documents"] = int(row["docs"] or 0)
|
||
|
|
summary["total_words"] = int(row["words"] or 0)
|
||
|
|
summary["avg_sentiment"] = round(float(row["avg_sent"] or 0), 3)
|
||
|
|
row = fetch_one("SELECT COUNT(DISTINCT lemma) AS c FROM document_word_counts WHERE NOT is_stopword")
|
||
|
|
summary["unique_words"] = int(row["c"] or 0) if row else 0
|
||
|
|
for label in ("positive", "neutral", "negative"):
|
||
|
|
row = fetch_one(
|
||
|
|
"SELECT COUNT(*) AS c FROM document_analytics WHERE sentiment_label = %s",
|
||
|
|
(label,),
|
||
|
|
)
|
||
|
|
summary[label] = int(row["c"] or 0) if row else 0
|
||
|
|
except Exception:
|
||
|
|
pass
|
||
|
|
|
||
|
|
top_words: list = []
|
||
|
|
documents: list = []
|
||
|
|
try:
|
||
|
|
top_words = fetch_all(
|
||
|
|
"""
|
||
|
|
SELECT lemma, MAX(token) AS token, SUM(count) AS total_count,
|
||
|
|
COUNT(DISTINCT storage_path) AS doc_count
|
||
|
|
FROM document_word_counts
|
||
|
|
WHERE NOT is_stopword
|
||
|
|
GROUP BY lemma
|
||
|
|
ORDER BY total_count DESC
|
||
|
|
LIMIT 30
|
||
|
|
"""
|
||
|
|
)
|
||
|
|
documents = _iso_rows(
|
||
|
|
fetch_all(
|
||
|
|
"""
|
||
|
|
SELECT filename, storage_path, doc_type, language, word_count,
|
||
|
|
unique_lemmas, sentiment_label, sentiment_compound,
|
||
|
|
sentiment_positive, sentiment_negative, sentiment_neutral,
|
||
|
|
extraction_method, analyzed_at,
|
||
|
|
COALESCE(user_labels, '{}') AS user_labels, label_notes, labeled_at
|
||
|
|
FROM document_analytics
|
||
|
|
ORDER BY analyzed_at DESC
|
||
|
|
LIMIT 50
|
||
|
|
"""
|
||
|
|
)
|
||
|
|
)
|
||
|
|
except Exception:
|
||
|
|
pass
|
||
|
|
|
||
|
|
return templates.TemplateResponse(
|
||
|
|
"documents.html",
|
||
|
|
{
|
||
|
|
"request": request,
|
||
|
|
"page_title": "Documents & Sentiment",
|
||
|
|
"summary": summary,
|
||
|
|
"top_words": top_words,
|
||
|
|
"documents": documents,
|
||
|
|
},
|
||
|
|
)
|